Wikipedia:Bots/Requests for approval/Qbugbot 4: Difference between revisions

Operator: Edibobb (talk · contribs · SUL · edit count · logs · page moves · block log · rights log · ANI search)

Time filed: 21:25, Tuesday, September 24, 2019 (UTC)

Function overview: This will resolve self-redirect links in arthropod articles by replacing the redirect page with a new article when possible (around 76% of the time) or by de-linking the self-redirect link.

Automatic, Supervised, or Manual: Automatic

Programming language(s): vb.net

Source code available: Yes. I will update User:Qbugbot/source before the first test.

Links to relevant discussions (where appropriate): https://en.wikipedia.org/wiki/Wikipedia_talk:WikiProject_Tree_of_Life/Archive_44#Request_for_comment%2C_qBugbot_and_self-redirects

Edit period(s): 8-24 hours per day

Estimated number of pages affected: 5,500 max

Namespace(s): Mainspace

Exclusion compliant (Yes/No): Yes

Function details: I would like to use qBugbot to resolve some self-redirects in arthropod articles. For clarity, and because it confuses me, when page A links to page B, and page B is a redirect page pointing back to page A, I am saying that page A contains a self-redirect link to page B.

About 585 arthropod articles contain self-redirect links pointing to 5428 redirect pages, excluding links to article subsections.

Here is what I am proposing:

• A self redirect link in a monotypic genus to its sole species will be de-linked in the genus page.
• Other self-redirects to species will have the species articles created to replace the redirect.
• Self-redirect links to genus pages will have an article created to replace the redirect page.
• Self-redirects to tribes, subfamilies, superfamilies, and higher ranks will be de-linked in general, but some of these articles may be created by the bot after manual review.
• Self-redirects to families will generally have the article created to replace the redirect, but these pages will be manually reviewed (113 cases).
• Self-redirect links in Automatic Taxaboxes and Speciesboxes will not be handled by the bot.

New articles can be created for about 4147 of 5428 redirect pages. The other 1253 or so are questionable for one reason or another. For them, the self redirect links will be de-linked and the redirect pages left unchanged.

A list of articles with self-redirect links is on the Qbugbot talk page.

{{BAG assistance needed}}

• Approved for trial (50 edits). Please provide a link to the relevant contributions and/or diffs when the trial is complete. Seem to recall that this is a trusted op with a good handle on edits, but let's make sure there are no bugs (pun intended). Primefac (talk) 12:51, 12 October 2019 (UTC)[reply]

  Trial complete. No major problems were encountered. Most edits replaced redirect pages with new articles. One edit removed some links from an existing article, Acerella. I inadvertently did 51 edits instead of 50 due to a mental malfunction.

  In future runs, the bot will add WikiProject templates to talk pages of articles that replace redirect pages and do not already have the templates. (I forgot about that.) I'll add WikiProject templates to those pages in this trial run when the bot gets authorization. I didn't do it now because I'm not sure whether it would be considered additional edits.

  Bob Webster (talk) 16:25, 14 October 2019 (UTC)[reply]

{{BAG assistance needed}}

• Hi Edibobb, could you quickly clarify this for me? "the bot will add WikiProject templates to talk pages of articles" is not mentioned anywhere else in this BRFA. Are you requesting to modify this BRFA or are you going to file another one or? --TheSandDoctor ^Talk 18:38, 26 October 2019 (UTC)[reply]

  When the bot creates an article, it also creates a talk page with something like {WikiProject Insects|class=stub|importance=low} in it. It occurred to me that this should also be done when an article replaces a redirect page. I had intended to do it now, as a slight modification (or clarification) to the BRFA, but if you think it would be better it's no problem to do this as a separate bot run. --Bob Webster (talk) 22:48, 26 October 2019 (UTC)[reply]

{{BAG assistance needed}}

@Edibobb: Do you have a list of diffs for review? Headbomb {t · c · p · b} 23:48, 1 November 2019 (UTC)[reply]

I just added one to the Qbugbot talk page. Bob Webster (talk) 06:36, 2 November 2019 (UTC)[reply]

Are we sure this is a good idea? I mean, this [1] doesn't contain any information beyond what the genus level Acerentomon article had. Headbomb {t · c · p · b} 17:30, 2 November 2019 (UTC)[reply]

It seems to me like the self-redirects should be eliminated either by un-linking the links to self-redirects or by replacing the redirect pages with stubs. Using the stubs does duplicate some information, but it also has some advantages:

• In some cases some additional information is provided in the new stub, such as the distribution range in Acerentomon noseki, occasional photos, IUCN status, etc.
• Replacing the self-redirects with stubs makes it easier for other editors to make more complete species pages.
• It many cases, such as this one, the new articles have better references.

Bob Webster (talk) 18:28, 2 November 2019 (UTC)[reply]

A couple of other thoughts: the TOL discussion seemed in favor of this project, and about 1000 of the redirect pages are for Genus rank instead of Species.

Bob Webster (talk) 05:54, 3 November 2019 (UTC)[reply]

{{BAG assistance needed}}

I've been traveling a bit in the past week, any BAG member can feel free to pick up this, otherwise I'll look at it once I'm back home next week. Headbomb {t · c · p · b} 18:10, 9 November 2019 (UTC)[reply]

Headbomb Is there anything I can do to speed the approval process? Bob Webster (talk) 20:30, 22 November 2019 (UTC)[reply]

@Edibobb: Can you provide a bit more information about what sources will be used for the created stubs? During previous runs of Qbugbot, there was some dissatisfaction with the references and further reading entries. Take [2] for example. It references Catalog of Life, ITIS, and World Spider Catalog, although Catalog of Life and ITIS get all of their spider data from World Spider Catalog, and are thus completely redundant. Also, BugGuide is a self-published (or community-published source), similar to Wikipedia, and probably doesn't qualify as a reliable source. Finally, only 1 of the 8 sources listed in the Further reading section have any mention of the topic species whatsoever. They seem to be related to the genus and family, not the species. I'm hoping that some of these issues can be fixed before we do the full run (if they haven't already been fixed). Kaldari (talk) 05:59, 4 December 2019 (UTC)[reply]

The references in Qbugbot were overhauled before the previous run (Qbugbut3), both in format and inclusion criteria. The result is far fewer references in "Further reading" and somewhat fewer references in "References", as well as more accurate and consistent references.

One of the main reasons for Qbugbot3 was to update the references from the previous run. In fact, the page you referred to is a good example. Your link goes the page originally created by Qbugbot in May, 2018. Last September, the page was updated by Qbugbot. The Catalogue of Life reference was removed. Six of the eight "Further reading" references were removed. There is one left that doesn't include specific information on the species, but I think it's worth including for the species of this genus.

World Spider Catalog and several "Species File" sites are used as references, when applicable, because they are the most complete and up-to-date authorities in their areas.

ITIS is used as a reference if it contains information about the article subject, because the distribution range, common names, synonyms, minor ranks, and other information are often obtained there.

GBIF is used as a reference if it contains information about the article subject, because it is the most complete and accurate general taxonomic reference. Even so, it can be a little out of date sometimes and has limited data for animals from some parts of the world.

Bugguide is used for a reference if it contains information about the article subject, because it usually has relevant information and photos not included in the Wikipedia article, supports the information in the Wikipedia article, and provides additional references.

Encyclopedia of Life is no longer used as a reference by Qbugbot. Catalogue of Life is not normally used as a reference because it rarely includes information not found in ITIS or GBIF.

ITIS, Bugguide, and Species File sites include ranks such as Tribe, Subfamily, Superfamily, and Suborder in their taxonomy. GBIF and World Spider Catalog do not.

Bugguide is published by Iowa State University. The species pages referred to by Qbugbot are created and maintained only by qualified editors, not by the community at large. GBIF.org now uses Bugguide as a source. There are more than 37,000 pages on en.wikipedia that refer to bugguide.

Bob Webster (talk) 16:08, 4 December 2019 (UTC)[reply]

@Edibobb: Thanks for the detailed reply! I'm glad to hear that most of these issues have been fixed! I'm also glad to hear that you aren't using Encyclopedia of Life and Catalog of Life as sources anymore as they are basically just tertiary sources that aggregate other sources. I'm still a bit skeptical of citing BugGuide (probably because I'm an editor there myself), but it's probably borderline enough to pass. Thanks for your hard work on this. Kaldari (talk) 19:31, 4 December 2019 (UTC)[reply]

I'm busier than expected, but I hope to be reviewing this by the end of the week. Any other BAG member can feel free to step in though. Headbomb {t · c · p · b} 21:52, 4 December 2019 (UTC)[reply]

@Edibobb: I notice these sort of follow up edits [3]. Could the bot handle those from the get go? Headbomb {t · c · p · b} 22:11, 4 December 2019 (UTC)[reply]

I've looked at that in the past. It would be easy enough to do, but I have not been able to find the categories to use. For example, should it be "Animals described in", "Insects described in", "Beetles described in", or "Scarabs described in"? It was discussed once in TOL, and at that time there wasn't a good way to determine the categories. Bob Webster (talk) 23:33, 4 December 2019 (UTC)[reply]

What about creating a new Category:Arthropods discovered in ... category three, of which the above could be subcategories of? Headbomb {t · c · p · b} 00:14, 5 December 2019 (UTC)[reply]

I looked into this. About a year ago (and maybe other times) a bunch of the "described in" categories were removed. Now the bot only needs to handle Animals, Crustaceans, Spiders, Insects, Beetles, Moths, and Butterflies. I've added it to Qbugbot. Bob Webster (talk) 04:55, 5 December 2019 (UTC)[reply]

Let's do another Approved for extended trial (10 edits). Please provide a link to the relevant contributions and/or diffs when the trial is complete.. Please notify the bug/entomology project and ask to comment on the addition of categories. Headbomb {t · c · p · b} 01:05, 6 December 2019 (UTC)[reply]

Trial complete.

I found no problems. For new articles that replace redirects, I added a talk page and the categories "Articles created by Qbugbot" and "Insects (or whatever) described in 1926 (or whenever)". There has been no response in the TOL project on the question of the "described in" categories, but the "Animals described in" category includes the ones I mentioned, and no other arthropods.

Here are the diff links:

• Acerpenna akataleptos
• Acerentulus americanus
• Acerentulus alpinus
• Acerentulus alni
• Acerentuloides
• Acerentomon tuxeni
• Actia brunnea
• Actinochaetopteryx nubifera
• Actenobius magneoculus
• Acontista vitrea

Bob Webster (talk) 19:22, 7 December 2019 (UTC)[reply]

{{BAG assistance needed}}

@Edibobb: link to TOL discussion? Headbomb {t · c · p · b} 04:56, 14 December 2019 (UTC)[reply]